Font Size: a A A

Reaserch On Multi-sensor Based Intelligent Transportation Vehicle Trajectory Optimization Method

Posted on:2020-07-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:J Y RuFull Text:PDF
GTID:1482306353963279Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
With the continuous development of sensor technology,network technology,and artificial intelligence technology,the multi-sensor intelligent transportation system with sunshine information acquisition and processing function has attracted extensive attention of researchers at home and abroad.The system has a wide range of application values in the fields of new energy utilization,smart city,smart travel,and safe driving.Based on the analysis and summary of relevant research at home and abroad,in-depth research on the optimal deployment of information gathering nodes and the analysis and optimization of vehicle trajectory's sunshine information in multi-sensor intelligent transportation system is made by this dissertation.The main research contents and results are as follows:Aiming at the problem of non-line-of-sight(NLOS)propagation of signals caused by obstacles in the optimal deployment of multi-sensor intelligent transportation system,a three-dimensional NLOS localization method based on improved hidden Markov model and interactive multi-model is proposed.According to the signal backtracking characteristics of nodes,the localization information of nodes is revised,and a NLOS localization algorithm based on improved hidden Markov model is proposed.The node motion model is divided into an omnidirectional model and orientation model,and the probability of the node in two states is evaluated by the interactive multi-model.The localization results of the improved hidden Markov model are fused,which effectively weakens the interference caused by NLOS error and object motion,and improves the positioning accuracy of the nodes.In order to solve the problem of coverage holes in the process of optimal deployment of multi-sensor intelligent transportation system,a repair algorithm based on improved cuckoo search coverage vulnerability is proposed.A three-dimensional perception model based on UAV sensor is constructed.The model considers the complex terrain and occlusion.It can accurately calculate and evaluate the complex environment area covered by directed sensors and establishes the criteria for judging the three-dimensional network coverage of the model.According to the coverage characteristics of directed sensor networks,an improved cuckoo optimization search algorithm is proposed,which achieves rapid repair of area coverage vulnerabilities.In order to solve the problem that there are a lot of redundant data in the vehicle trajectory analysis and optimization process of multi-sensor intelligent transportation system,a simplified analysis method of vehicle trajectory based on sunshine information is proposed.Based on the sunshine information provided by the multi-sensor intelligent transportation system and the angular characteristics of the trajectory,a binary direction model for abstracting the sunshine information of the trajectory is proposed.The single trajectory simplification problem is transformed into a network problem,and the shortest path algorithm is used to obtain the minimum information points that need to in the vehicle trajectory.Experiments on vehicle single trajectory simplification using Minnesota traffic data show that the algorithm can effectively reduce redundancy points in data and improve the efficiency of the system.Aiming at the redundancy of multi-trajectory information in vehicle trajectory sunshine Information Query in the process of vehicle trajectory analysis and optimization in multi-sensor intelligent transportation system,a trajectory compression algorithm based on sunshine query information is proposed.The algorithm utilizes the trajectory similarity of the road network to optimize the calculation method of sunshine time error.The trajectory is defined as dominant trajectory and dependent trajectory,and a dominant-dependent trajectory matching similarity model based on sunshine query information trajectory similarity is established.The optimal matching is selected according to the minimal dominant set.The trajectory compression is achieved by reconstructing the trajectory data.A sunshine information query platform for ITS is developed and built,which can test,analyze,and apply sunshine information of vehicle trajectory.
Keywords/Search Tags:Wireless sensor network localization, network coverage, sunshine information, intelligent transportation system, trajectory simplification, trajectory compression
PDF Full Text Request
Related items